Chapter  37:  Management of Dental Patients Who Are HIV-Positive: Evidence Report/Technology Assessment Number 37

Search Process

Two automated databases were searched -- MEDLINE and EMBASE -- and the contents of the Cochrane Collaboration Library were reviewed manually. We also hand-searched the contents of the most recent 12 months (through spring 2000) of the five journals most likely to contain relevant articles -- Oral Surgery, Oral Medicine, Oral Pathology, Oral Radiology and Endodontics; Journal of Oral Pathology and Medicine; Oral Diseases; AIDS; and Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology. Our review of the Cochrane materials indicated related topics that were under study but not currently available as published reports. No search was made through the "gray" literature (i.e., unpublished materials).

Separate MEDLINE searches were conducted for each of the five key questions, but all five searches began with the terms "HIV" or "HIV infection" or "acquired immunodeficiency syndrome" and were limited to humans and the English language. Beyond these similarities, the searches were customized except that the search terms for Key Questions 2A and 2B were similar to each other up to a point, as were the search terms for Key Questions 3A and 3B. We did the searches without restriction on the year of publication, but we found no relevant publications dated before 1983. All five of the MEDLINE searches started with 83,962 articles on human HIV or AIDS in the English language. Inclusion and exclusion criteria used by the reviewers are set out in the next subsection.

For Key Question 1, the search then included a list of dental procedures of interest -- dental care, tooth extraction, gingivoplasty, periodontitis, periodontal surgery, dental implants, gingivectomy, oral surgical procedures, orthognathic surgery, dental scaling, dental prophylaxis, root scaling, root planing, root canal therapy, and pulpectomy. We identified 145,292 articles on these topics. When cross-checked against the English-language articles on HIV/AIDS, 767 articles were common to both search results. These were further winnowed down to 201 articles by eliminating nonresearch articles and clinical case reports. We included only studies described in such terms as controlled clinical trials, randomized controlled trials, multicenter study, epidemiologic research design, comparative study, evaluation study, outcome and process assessment, outcome assessment, or treatment outcome.

For Key Question 2A, the departure point from the first search was to specify the oral conditions of interest -- mouth diseases, periodontal diseases, oral infections, hairy leukoplakia, oral leukoplakia, oral candidiasis, necrotizing ulcerative gingivitis, oral ulcers, parotid diseases, parotid gland, parotitis, or parotid swelling. This resulted in 1,719 articles when cross-checked against the English-language articles on HIV/AIDS. These were reduced further to 297 articles by restricting them to mention of HIV seropositivity, seroconversion, or AIDS serodiagnosis.

The search for information regarding Key Question 2B included searching for the same oral conditions as in Key Question 2A, but we restricted the studies to those mentioning immunosuppression or CD4 lymphocyte count. This yielded 133 articles.

For Key Question 3A, the search was limited to oral candidiasis and to a list of antifungal agents -- fluconazole, ketoconazole, nystatin, itraconazole, clotrimazole, amphotericin B, or drug therapy. This resulted in 223 articles, a number that was further reduced to 18 by restricting the search with the terms primary prevention, prevention, preventive medicine, health promotion, disease prevention, dental prophylaxis, or prophylaxis.

The search strategy for Key Question 3B was similar to the search done for Key Question 3A to the point of specifying the condition and the drugs. We then restricted the articles to those dealing with drug therapy, therapy, intervention studies, intervention, treatment outcome, or treatment, which yielded 143 articles.

Subsequent to the MEDLINE searches, we conducted additional searches on EMBASE. EMBASE is somewhat more limited than MEDLINE for the time period covered (only since 1988) and in the search terms available to customize queries. For this latter reason, we did only three customized searches for the five questions: Key Questions 2A and 2B were combined into a single search, as were Key Questions 3A and 3B. The vast majority of articles uncovered in these searches were duplicative of those found in the MEDLINE searches. The EMBASE searches resulted in the addition of only three articles to the bibliography.

Inclusion/Exclusion Criteria

The basic screening criteria for inclusion and exclusion that applied to articles for all of the questions were as follows:

• Include: Only research manuscripts published in English reporting on the human population.

• Exclude: Literature reviews (except to check references for additional articles), letters, commentaries, editorials, clinical case reports, or practice or treatment guidelines.

For Key Question 1 on the differences in outcomes and complications of intra-oral procedures between HIV-positive and HIV-negative patients, the criteria used were as follows:

• Include: Studies reporting on complications of intra-oral surgical dental procedures, including orthognathic, periodontal, extractions, endodontics, prophylaxis, scaling and root planing, and implants. The complications reported for both HIV-positive and HIV-negative patients are local infection, systemic infection, increased bleeding, dry socket (alveolitis), and delayed healing.

• Exclude: Studies in which treatment is not rendered concurrently to HIV-positive and HIV-negative patients or complications are not reported according to patient group and procedure.

For Key Question 2A on the use of oral lesions as markers for seroconversion, the criteria were as follows:

• Include: Reported presence of any of the following selected oral lesions -- hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, or parotid swelling -- within the 3-month window for primary HIV infection; reporting of the specificity, sensitivity, or positive or negative predictive values of the selected oral lesions as markers for seroconversion; or information from which these values can be calculated.

• Exclude: No reasonable or firmly established time of exposure to HIV.

The criteria for Key Question 2B addressing the use of selected oral lesions as indicators of change in immunosuppression were as follows:

• Include: Presence of any of the following oral lesions -- oral hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, parotid swelling, Kaposi's sarcoma, or linear gingival erythema -- in HIV-positive patients; the HIV status of the patients at the time of the oral lesion is reported by CD4 cell count or plasma viral load; and reporting of the specificity, sensitivity, or positive or negative predictive values of the selected oral lesions as indicators of severe immunosuppression (CD4 <200 cells/mm³); or information from which these values can be calculated.

For Key Question 3A, addressing the use of available antifungal agents to prevent initial or recurrent oral candidiasis, the criteria were as follows:

• Include: Studies of the following available antifungal agents -- nystatin formulations, clotrimazole, amphotericin B suspension, ketoconazole, fluconazole formulations, or itraconazole formulations; studies of the prevention of initial infection or recurrence of oropharyngeal candidiasis; and studies whose results are reported separately for initial infection or recurrence.

• Exclude: Studies on non-HIV/AIDS patients only or studies not reporting results separately for HIV/AIDS patients, studies without laboratory confirmation of oral candidiasis status at start of study period and after apparent infection, studies of mixed-site candidiasis where results are not reported separately for oropharyngeal candidiasis, or studies without a concurrent control or comparison group.

The criteria for Key Question 3B, on the effectiveness of available antifungal agents to treat oral candidiasis, were as follows:

• Include: Studies of available antifungal agents -- nystatin formulations, clotrimazole, amphotericin B suspension, ketoconazole, fluconazole formulations, and itraconazole formulations -- and their use in treatment of oropharyngeal candidiasis.

• Exclude: Studies without a concurrent control or comparison group, studies comparing different formulations of the same drug, studies on non-HIV/AIDS patients only or studies not reporting results separately for HIV/AIDS patients, studies without laboratory confirmation of oral candidiasis status at time of diagnosis and after treatment, or studies of mixed-site candidiasis where results are not reported separately for oropharyngeal candidiasis.

Reviewing and Abstracting the Articles

Determining which articles identified in the literature searches were to be included in the review involved a number of steps. First, the titles of articles for each of the questions received independent dual review by two of the authors, and copies were obtained of any abstracts suggested by either one of them. Then copies of the abstracts were reviewed by two of the authors. Articles were identified for abstraction only when both reviewers agreed. The abstracts of all of the excluded articles were independently reviewed by another of the authors to ensure that no articles were incorrectly excluded from the review. Eligible articles were given to one of three trained abstractors, who extracted the data to be included in the evidence tables using specially designed forms. Authors then checked the abstracted data against the articles as they prepared their sections of the report and the corresponding evidence tables.

Complications of Oral Procedures

The data from five articles were abstracted for the review of Key Question 1, comparing the complication rates associated with dental treatment in HIV-positive persons with the complication rates for similar dental treatment in individuals who are HIV negative. Only two of the seven dental procedures specifically mentioned in Key Question 1 were the objects of study in the five articles. One article examined endodontic procedures (root canal treatment), and four studies examined tooth extraction. None reported on orthognathic surgery, periodontal surgery, prophylaxis, scaling and root planing, or implants.

• In the study of endodontics, the immediate (1- to 3-month follow-up) postoperative complication rate was exceedingly low in the HIV-positive group and nonexistent in the control group. Only one of 48 patients experienced any postoperative complications; an asymptomatic HIV-positive male was found to have pain and swelling following the initial root canal treatment. He received local debridement and antibiotics, and no further complications occurred. Otherwise, no complications associated with endodontic therapy were noted in any of the patients regardless of whether they did or did not receive prophylactic antibiotics. The authors did not detect a clinically significant difference in complication rates between the two groups.

• Three of the four studies in the review of dental extractions found no statistically significant difference in complications between the HIV-positive and HIV-negative groups, although the HIV-positive groups tended to have more postoperative complications. The final study found that the HIV-positive groups had a statistically significantly higher complication rate, but with adjustment for risk factors, the difference was no longer significant. Postextraction complications included persistent bleeding, persistent pain, localized alveolitis, local wound infection, or delayed wound healing. Nevertheless, across all studies, the postoperative complications that were experienced were rather minor, and when they occurred they were treated on an outpatient basis. Finally, based on their findings, none of the four studies called for the need to take special precautions with HIV-positive patients who do not have a coagulopathy (e.g., hemophilia, thrombocytopenia, or other known bleeding disorders) and are sufficiently healthy to be seen on an outpatient basis.

Oral Conditions as Markers of Seroconversion

Only one study was abstracted in the review for Key Question 2A. The question asked for evidence on a set of specific oral conditions -- hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, and parotid swelling -- as markers for recent HIV seroconversion. The study investigated the prevalence of a variety of signs and symptoms (including oral candidiasis) among two groups of initially seronegative hospital patients. All of the patients received blood transfusions for a variety of reasons, but half were transfused with seropositive blood while the "matched controls" received seronegative blood.

The question uses the medical "testing" dimensions of sensitivity (Sn), specificity (Sp), positive predictive value (PPV), and negative predictive value (NPV) of the oral conditions as a marker of recent seroconversion as the way of evaluating their usefulness. Positive enzyme-linked immunosorbent assays (ELISAs) and Western blot tests for HIV, obtained from patients previously confirmed to be seronegative for up to 3 months after the presumed time of infection/exposure, were considered evidence of recent seroconversion. No special conditions were imposed on the diagnosis of the oral conditions.

• The majority of patients with oral candidiasis had seroconverted by the end of 3 months (PPV = 82 percent), but only a small proportion of those who had seroconverted had oral candidiasis (Sn = 14 percent). Very few who did not seroconvert had oral candidiasis (Sp = 97 percent), and most of those who did not have oral candidiasis did not seroconvert either (NPV = 57 percent).

Oral Conditions as Indicators of Severe Immunosuppression

Key Question 2B involved assessing evidence supporting the use of selected oral conditions as indicators of the progression of HIV infection to the stage of severe immunosuppression and a diagnosis of AIDS. The question suggests the use of the CD4 lymphocyte cell count as a measure of immunosuppression. The question identified the following seven oral conditions of interest: hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, parotid swelling, linear gingival erythema, and Kaposi's sarcoma. The data from 10 articles were abstracted and included in the review for this question. All 10 articles reported CD4 count alone as their measure of immunosuppression.

This question also uses the medical "testing" dimensions (Sn, Sp, PPV, and NPV) of the oral conditions to evaluate their usefulness as indicators of immunosuppression. In this context, an oral condition with high sensitivity would represent a screening tool for identifying HIV disease that has progressed to the next stage. When sensitivity is high, it means that a large proportion of the persons whose HIV disease has progressed to the next stage also have the particular oral condition. Thus, when you find the oral condition, you are likely to find a more advanced stage of HIV. High specificity, on the other hand, indicates that a large proportion of persons whose HIV disease has not progressed to the next level also do not have the particular oral condition. In the absence of high Sn, high Sp is not particularly useful. High PPV, on the other hand, is a measure that represents how closely the presence of the oral condition is associated with the progress of HIV to the next stage. When PPV is high, it indicates that a large proportion of the persons who clinically present with the oral condition also have HIV disease that has progressed to the next stage. A high NPV, on the other hand, indicates that a large proportion of those without the oral condition also do not have HIV that has progressed to the next stage. In the absence of a high PPV, high NPV is not particularly useful.

The oral conditions on which studies reported were quite varied. None of the articles reported on parotid swelling; however, all 10 articles reported on oral candidiasis, six reported on hairy leukoplakia, and four on oral ulcers. Only two studies reported on linear gingival erythema, two on necrotizing ulcerative periodontitis, and three on Kaposi's sarcoma.

• There is little apparent consistency in Sns across the 10 studies of oral candidiasis. They range from 20 to 77 percent, with a small cluster of five study groups in the lower end of the range (from 20 to 41 percent). The situation with respect to Sps is somewhat more consistent. They range from 65 to 97 percent, with a cluster of eight study groups in the upper end of the range (from 79 to 97 percent). The PPVs range from 34 to 88 percent, with eight clustered at the low end of the range (from 34 to 58 percent). The range of NPVs was from 61 to 90 percent; however, all but four of the NPVs were clustered between 84 and 90 percent.

• Across the six studies of hairy leukoplakia, Sns and Sps were fairly consistent. The Sns ranged from 13 to 24 percent and the Sps ranged from 83 to 95 percent. PPVs ranged from 29 to 70 percent, with four below 50 percent, and NPVs ranged from 51 to 86 percent.

• The Sns reported in the two studies of necrotizing ulcerative periodontitis were very low, ranging from 0 to 16 percent, whereas the Sps were very high (92 to 99 percent). With the exception of one study group that had no cases of this condition, the PPVs ranged from 80 to 95 percent. The NPVs ranged from 52 to 70 percent.

• In the four studies examining oral ulcers, Sns ranged from 0 to 21 percent, while the Sps ranged from 88 to 100 percent. PPVs ranged from 0 to 100 percent, and NPVs ranged from 52 to 84 percent.

• The two studies of linear gingival erythema had low Sns (13 to 34 percent) but relatively high Sps (62 to 90 percent). The PPVs ranged from 18 to 55 percent, while the NPVs ranged from 61 to 82 percent.

• Sns across the three studies of Kaposi's sarcoma were low and ranged from 7 to 12 percent. However, the Sps were extremely high, ranging from 99 to 100 percent. The PPVs were high as well, ranging from 73 to 100 percent. Also, the NPVs were reasonably high, ranging from 61 to 84 percent.

Efficacy of Antifungals as Prophylaxis for Oral Candidiasis

There were six clinical trials that examined the efficacy of available antifungals to prevent oral candidiasis in persons who are HIV positive. Only two of the six available antifungals were studied in this context. There were five studies of fluconazole and one of nystatin pastilles.

• Fluconazole was statistically significantly more efficacious than was placebo in preventing recurrences or new infections over 3 to 17 months, when studied at doses ranging from 100 mg/day to 100 mg/week. Gastrointestinal disorders were the most common, but tolerable, side effects.

• Nystatin pastilles at 200,000 units/day and 400,000 units/day were efficacious at preventing new or recurrent oropharyngeal candidal infections, with the higher dose being more effective. Few side effects were reported.

Effectiveness of Antifungals as Treatment for Oral Candidiasis

Twelve clinical trials evaluated the effectiveness of the six available antifungal drugs in the treatment of oral candidiasis in persons with HIV/AIDS. No studies included an evaluation of amphotericin B suspension, and fluconazole was the most studied of the five remaining.

• Fluconazole appears to be from 88 to 100 percent effective in obtaining a complete clinical response after 14 days of therapy and from 53 to 76 percent effective in obtaining a culture negative for Candida species.

• Itraconazole appears to be roughly equivalent in effectiveness to fluconazole, with ketoconazole achieving the same or slightly lower response rates.

• Fluconazole and itraconazole are more effective at managing oropharyngeal candidiasis than are nystatin or clotrimazole, particularly when mycological response rates and relapse rates are taken into account.

Key Clinical Questions

The clinical questions to be addressed in this evidence report are based on the three issues mentioned briefly above. The actual questions were proposed by the NIDCR staff involved in the planning of the State of the Science Workshop on the management of HIV-positive dental patients. We subsequently refined the questions through discussions with the NIDCR staff and further clarified them through communication with the Technical Expert Advisory Group (TEAG) assembled for this particular topic area (see below). The questions reflect two common concerns: (1) that healthcare practitioners, and particularly dentists, may not be aware of the available research with respect to the treatment of persons who are HIV positive, and (2) that the research may not be as comprehensive or definitive as it should be. The questions cover three issues: (1) risks to HIV-positive patients of invasive oral procedures, (2) specific oral conditions as markers of change in HIV immune status, and (3) antifungal treatment of oral candidiasis in persons with HIV/AIDS. We then split the last two issues in two, so we actually address five questions in this report. (The five questions are discussed in greater detail in Chapter 2.)

The first main question addresses whether persons who are HIV positive run any greater risk of complications from selected invasive surgical dental procedures than do persons who are not HIV positive. The underlying issue concerns whether special precautions or measures -- expertise, facilities, medications -- not needed for seronegative dental patients are necessary to protect patients who are seropositive against such possible complications as systemic infection, delayed healing, or excessive bleeding because of their immunocompromised state. The answer to this question could certainly have an impact on the availability of routine dental services to persons with HIV/AIDS.

The second main question involves the usefulness and validity of using certain oral conditions commonly reported among persons with HIV/AIDS as markers or indicators of important changes in immune status. Two critical points, each represented by separate subquestions, come to the fore in this context. The first asks whether these oral conditions might be used as markers of the stage of seroconversion from HIV negative to positive. Some HIV researchers have suggested that during the period of seroconversion, the infected person is especially vulnerable to these oral conditions. Under these circumstances, the presence of the condition could be an indicator that HIV testing is warranted and that early intervention therapy might be justified. The second critical point occurs when HIV-infected persons become sufficiently immunocompromised to meet the definition of symptomatic AIDS. Again, experts have suggested that at this juncture of the infection, the person is particularly susceptible to these oral conditions. If so, and if the presence of the oral conditions is associated with this change in immune status as measured by available measures of the immune system (CD4 count) or level of viral load, it may be a signal to initiate different antiretroviral therapy.

The third main question concerns the effectiveness of available antifungal agents in the prevention and treatment of oral candidiasis in persons who are HIV positive. This question is actually two separate questions because we have no reason to assume that a given agent will be as effective in a prophylactic mode (preventing either initial or recurrent infection) as in a treatment capacity. The outcomes of interest, or end points, are quite different as well. In either case, side effects of the antifungal agents are a relevant concern. Moreover, the use of antifungal agents, especially in a prophylactic mode, raises the issue of increasing resistant strains of Candida species.

Technical Expert Advisory Group (TEAG)

Guidelines from AHRQ required identification of technical experts in the specialized area of managing the oral conditions of dental patients with HIV/AIDS to serve as report advisers. A TEAG is expected to contribute to advancing AHRQ's broader goals of (a) creating and maintaining science partnerships and public-private partnerships and (b) meeting the needs of an array of potential consumers and users of its products. Thus, a TEAG is both an additional resource and a sounding board throughout the project.

Our TEAG consisted of six individuals who are acknowledged technical or clinical experts in this area. One member was nominated to serve as a representative of a patient advocacy group (the National Association of Persons with AIDS [NAPWA]), and another was chosen to represent the perspective of a potential user group (the American Dental Association [ADA]). Refer to Appendix B for a more complete description of the selection, role, and composition of the TEAG.

To ensure scientifically robust work, we asked the TEAG to provide reactions to work in progress and to advise on substantive issues or possibly overlooked areas of research. TEAG members participated in conference calls and e-mail communications at the beginning of the project to discuss the key clinical questions, initial drafts of causal pathways, and proposed inclusion and exclusion criteria for research articles; during the development of article abstraction forms to provide comments on the forms, the content proposed for inclusion in the evidence tables, and the final versions of the key clinical questions and causal pathways; and when the draft evidence tables were produced to discuss the content of the tables and the completeness of the search.

Because of their extensive knowledge of the literature and ongoing research in this specialized area of dentistry dealing with the management of patients who are HIV positive, as well as their active involvement in the associated professional societies, we also asked TEAG members to participate in the peer review process by commenting on the draft evidence report.

Prevalence of HIV Infection and Oral Diseases in Patients with HIV/AIDS

The true prevalence of HIV infection in the United States is unknown because HIV case reporting through the CDC national surveillance mechanism is not universal among States and because not all HIV-infected individuals have been tested. By contrast, reporting of AIDS cases is required in the United States. Through the end of 1999, 733,374 persons had been reported with AIDS, and from the 34 States that report cases of HIV infections an additional 113,167 persons were determined to be living with diagnosed HIV infection (not AIDS).¹³

Oral mucosal opportunistic infections are common among individuals with HIV/AIDS, occurring in an estimated 20 to 50 percent of HIV-infected patients during the course of their disease.^14,15 Oral candidiasis and oral hairy leukoplakia have been the most common oral mucosal lesions associated with HIV disease and may be markers of immunosuppression and/or disease progression.^16-18 There is some evidence, however, to suggest that the success of new AIDS treatment regimens (HAART) may be changing the incidence and prevalence of oral diseases among persons with AIDS.^19-21 Oral lesions may also give clues to primary acute HIV infection.^22-24

In addition to high levels of chronic gingivitis and adult periodontitis,^25-27 individuals with HIV are susceptible to rare and severe forms of necrotizing ulcerative gingival and periodontal breakdown.²⁸ Compared with healthy adults, HIV-infected individuals have more prevalent, severe, and extensive conventional gingivitis and higher levels of periodontal pocketing and attachment loss, two clinical indicators of adult periodontitis.²⁹ In a cross-sectional, masked study, children with perinatally acquired HIV were shown to be at greater risk for caries than their siblings, more so with advancing disease.³⁰ A comparison of caries experience between HIV-positive and general patients did not reveal statistically significant differences between them.³¹ Also not yet documented are the prevalence, extent, and quality-of-life effects of xerostomia resulting either as a side effect of medication or HIV-induced salivary gland dysfunction, and the impact of this condition on dental caries rates among persons who are HIV positive.

Burden of HIV Infection and Oral Disease in HIV Patients

The burden of newly diagnosed HIV/AIDS illness falls increasingly upon racial and ethnic minorities, women, adolescents, and the financially disadvantaged in the United States, and decreasingly upon men who have sex with men (MSM), although this group remains the largest single exposure group.¹³ Incident AIDS cases increasingly represent people who are not diagnosed until they develop opportunistic infections, those who do not obtain medical treatment once diagnosed, and those for whom antiretroviral drug treatments fail.¹³ Although AIDS incidence rates have been decreasing generally, these decreases are smaller among women, blacks, people infected through heterosexual contact, and those living in the South. However, there is recent evidence that AIDS incidence is actually increasing among young homosexual/bisexual men. Additionally, with the new HAART treatment, there has been improved survival and more limited disease progression to AIDS, resulting in more persons living with HIV as a chronic disease.

There are disparities in perceived dental needs and access to dental care among those who are HIV infected. A 1988 - 89 survey of individuals with symptomatic HIV disease in nine U.S. cities revealed a stronger perceived need for dental care (52 percent) than for any other service.³² Multivariate analysis showed that people who were white, were in low-income groups, used intravenous drugs, or had a history of oral opportunistic infections were more likely to report dental need. Forty-seven percent reported having an opportunistic oral infection, and these were significantly more prevalent among whites and the severely ill. In a 1994 multi-State interview of people who reported to local and State health departments through CDC's HIV/AIDS surveillance, 32 percent were not currently receiving dental care although dental needs were high, including 34 percent reporting current toothaches and 39 percent having painful or bleeding gums.³³ A multivariate model of factors associated with persons who received dental care after they learned of HIV-positive status was significant in that these persons had dental insurance, 12 or more years of education, two or more oral conditions, and were white.³³

Two recent reports from the 1996 HIV Cost and Services Utilization Study cohort, a nationally representative probability sample of 2,864 HIV-infected persons receiving medical care, have added further insight into dental need and utilization.^10,11 Among subjects, 19.3 percent had a perceived unmet need for dental care in the previous 6 months.¹¹ Unmet dental need was highest among those with no dental insurance, those on Medicaid in States without dental benefits, those with incomes under $5,000, and patients with less than a high school education.¹¹ Forty-two percent of the sample had seen a dental health professional in the preceding 6 months; however, African Americans, persons whose exposure to HIV was via blood products, persons with less education, and those who were unemployed were less likely to get dental care.¹⁰ Availability of a regular source of care was associated with increased utilization.

Oral mucosal and periodontal diseases can result in morbidity and affect the quality of life for the HIV-infected patient.³¹ In a comparison of dental patients with HIV infection and general dental patients receiving publicly funded care in Adelaide, South Australia, HIV patients reported significantly greater levels of social impact. Social impact among HIV patients was frequent: 64.6 percent reported toothache, 43.7 percent avoided eating, and 16.7 percent avoided going out because of dental problems.³¹ Although the index of decayed, missing, and filled permanent teeth (DMFT) and its components did not differ significantly between HIV-infected and general dental patients, oral candidiasis was present among 32 percent of HIV patients.³¹

The most typical presentation of oral candidiasis is pseudomembraneous candidiasis, or thrush, which may be accompanied by pain during swallowing (odynophagia) that makes eating difficult, thus contributing to impaired nutritional intake and weight loss. Utilization of care for oral lesions, such as thrush and oral sores, as reported by participants in the national multisite AIDS Cost and Service Utilization Survey in 1991 - 92 was relatively low -- 9.1 percent in all -- and educational and racial differences existed among respondents who received care for HIV-associated oral lesions.³⁴

Variations in Practice for HIV-Positive Patients

Antifungal Drug Management and Prophylaxis of Oral Candidiasis

Table 1. FDA-Approved Antifungal Drugs Used for the (more...)

Table 1. FDA-Approved Antifungal Drugs Used for the Management of Oropharyngeal Candidiasis^a

Generic	Proprietary	Formulation b	Quantity	2000 AWP c /Typical Treatment Course
amphotericin B	Fungizone	100 mg/mL oral suspension	$70.56 / 56 mL	Rx: 100 mg QID for 14 days
clotrimazole	Mycelex	10 mg troche	$62.43 / 50	Rx: 10 mg 5x/day for 10 days
fluconazole	Diflucan	100 mg tablet	$112.03 / 15	Rx: 200 mg stat and 100 mg QD for 13 days
		10 mg/mL oral solution	$122.16 / 140 mL	Rx: 100 mg QD for 14 days
		40 mg/mL oral solution	$110.95 / 35 mL	Rx: 100 mg QD for 14 days
itraconazole	Sporanox	100 mg capsule	$198.00 / 28	Rx: 200 mg QD for 14 days
		10 mg/mL oral solution	$207.48 / 280 mL	Rx: 200 mg QD for 14 days
ketoconazole	Nizoral	200 mg tablet	$52.66 / 15	Rx: 400 mg stat and 200 mg QD for 13 days
nystatin	Mycostatin	100,000 U/mL oral suspension	$69.85 / 200 mL	Rx: 500,000 U QID for 10 days
		200,000 U/oral pastille	$52.33 / 50	Rx: 200,000 U 5x/day for 10 days
		500,000 U/oral tablet	$27.48 / 40	Rx: 500,000 U QID for 10 days

a

The following clarification is provided in a memo dated September 20, 2000, from Bernard A. Schwetz, D.V.M., Ph.D., acting deputy commissioner, U.S. Food and Drug Administration: "There are six antifungal drugs approved for marketing. These are fluconazole, itraconazole, clotrimazole, amphotericin B, nystatin, and ketoconazole. Of these, nystatin, itraconazole, and fluconazole are specifically indicated in approved labeling for treating oral candidiasis. Fluconazole (Diflucan®) is also indicated prophylactically 'to decrease the incidence of candidiasis in patients undergoing bone marrow transplantation who receive cytotoxic chemotherapy and/or radiation therapy.' Clotrimazole is indicated for treating candidiasis, but oral candidiasis is not specifically mentioned. Ketoconazole is approved for treating cutaneous candidiasis, and amphotericin B is not labeled for treating candidial infections."

b

mg/mL = milligrams per milliliter, U/mL = units per milliliter

c

AWP = average wholesale price

QID = four times a day, QD = every day, STAT = immediately

Source: Medical Economics. Drug Topics Red Book. Montvale, NJ: Medical Economics Co., Inc., 2000.

Clinicians use a variety of drugs, in both topical and systemic delivery systems, to manage fungal infections, including oral candidiasis. Table 1 lists the six drug preparations approved for marketing in the United States by the Food and Drug Administration (FDA) that are used to manage fungal infections, along with a description of a typical course of treatment for oral candidiasis and the average wholesale price associated with such a course of treatment. For some of these antifungals, use for treating oral candidiasis may represent an "off-label" use.

Several antifungal agents, including nystatin, itraconazole, ketoconazole, clotrimazole, and amphotericin B oral suspension, can be used topically. For topical agents, successful therapy depends on adequate contact time (2 minutes) between the agent and the oral mucosa. Treatment time varies from 7 to 14 days, with therapy minimally continued for 2 to 3 days beyond the last appearance of clinical signs and symptoms. Topical agents have the benefit of few side effects at normal therapeutic doses because of their lack of gastrointestinal absorption. However, sucrose-containing topical agents can be cariogenic when used over prolonged time periods, so adjunctive topical fluoride therapy may be needed.

Nystatin is a polyene antifungal administered topically. The general failure of the major fungal pathogen Candida albicans to develop resistance to nystatin, its negligible gastrointestinal absorption and thus relatively minor and infrequent side effects, and its topical mode of action make nystatin pastilles (lozenges) a potentially excellent choice for use as a prophylactic agent.⁵³ Whereas the nystatin oral suspension used to "swish and swallow" has a high sucrose content, the pastille formulation requires adequate saliva. Clotrimazole, a topical imidazole, also requires adequate saliva to dissolve the troches (lozenges), but they contain dextrose rather than sucrose as a sweetening agent.

Systemic antifungals, including ketoconazole, fluconazole, and itraconazole, have the advantage of once-daily dosing and treatment of fungal infections at multiple body sites at once. However, these antifungals have more side effects and require consideration of important drug interactions.

Ketoconazole, an imidazole, may interfere with the metabolism of cyclosporine, tacrolimus, and warfarin, increasing toxicity. Its use is also contraindicated with isoniazid, phenytoin, and rifampin because they decrease its effect. Terfenadine and astemizole are also contraindicated while taking ketoconazole. Concomitant use of ketoconazole with protease inhibitors, which are normally metabolized through the cytochrome P450-34A enzyme system on which ketoconazole acts, may produce increased levels of the protease inhibitors. Absorption of ketoconazole requires gastric acidity, and that may necessitate the concomitant use of antacids, H₂ antagonists, omeprazole, and sucralfate, which may reduce its bioavailability and result in treatment failure. Ketoconazole is to be taken with food. Periodic liver function tests are recommended to monitor for hepatotoxicity.

Fluconazole, a novel bis-triazole antifungal agent introduced in 1990, has systemic effects that may be beneficial for other fungal infections. Subjects in the fluconazole prophylactic arm of one antifungal placebo-controlled trial showed improvement of dermatophytoses, such as tinea pedis, onychomycosis, and tinea cruris.⁵⁴ Additionally, systemic fluconazole prophylaxis may prevent esophageal and vaginal candidiasis,⁵⁵ cryptococcemia, histoplasmosis, and other deep fungal infections. Unlike ketoconazole, fluconazole is not altered by changes in gastric acidity and carries less risk of hepatotoxicity; however, many of the same drug interactions are possible.

Itraconazole oral solution, also a triazole antifungal with systemic properties that is indicated for the treatment of oropharyngeal candidiasis, has demonstrated clinical improvement of fluconazole-refractory oropharyngeal candidiasis in HIV patients, allowing clinical response in 55 percent of patients in a median of 7 days.⁵⁶ It carries the same drug interaction potential and the common azole antifungal side effects of nausea, headache, and abdominal pain, as well as reduced absorption under acidic stomach conditions.

Of the antifungal agents used to treat oral candidiasis in people with HIV/AIDS, fluconazole has consistently shown excellent clinical response, limited drug interactions, and mild toxicities, making it the likely drug of choice and thus the most widely prescribed. A newly raised concern about the widespread use of fluconazole as chronic suppressive therapy for oral candidiasis is the potential for development of azole-resistant Candida albicans and selection of non-albicans Candida species, which also increase in prevalence with immune-system decline and thus further complicate management of some individuals.^57-59 Fluconazole resistance is most likely with substantial prior exposure to fluconazole and with severe immunosuppression (e.g., CD4 <50 cells/mm³).^60,61

Because oral candidiasis can be a frequent and significant source of oral discomfort, pain, loss of taste, and aversion to food, it may lead to secondary complications, such as esophageal candidiasis, for some patients with HIV/AIDS. For this reason, it may justify consideration of antifungal prophylaxis in such patients. Although rarely used for primary prevention, it is sometimes considered for patients prone to multiple recurrences of oral candidiasis. Improved quality of life resulting from longer symptom-free periods is important to many patients. The possibility that resistance to antifungal agents will develop and consideration of the cost of the drug must also be weighed in the decision to attempt prophylaxis and in the selection of an antifungal agent.

Patients with CD4 counts of less than 200 cells/mm³ may be most at risk for oral candidiasis and thus may benefit most from antifungal prophylaxis. Candida albicans carriage and history of oral candidiasis are other significant predictors of time to oral candidiasis.⁵³

Because of the effectiveness of antifungal therapy for the treatment of acute oropharyngeal candidiasis, the low mortality associated with mucosal candidiasis, the potential for resistant Candida organisms to develop, the possibility of drug interactions, and the cost of prophylaxis (e.g., annual average wholesale cost per patient of 200 mg of fluconazole per day in 1999 was $4,267), the U.S. Public Health Service and the Infectious Disease Society of America do not recommend routine primary prophylaxis.⁶² Expert opinion also recommends against chronic prophylaxis of recurrent oropharyngeal candidiasis for the same reason. When recurrences are frequent or severe, oral azoles (fluconazole, itraconazole, or ketoconazole) may be considered. Factors considered in decisions for prophylaxis against oral candidiasis include the following: the impact of oral fungal disease recurrences on the patient's quality of life and well-being, the need for prophylaxis for other fungal infections, cost, toxicities, drug interactions, and the potential to induce antifungal drug resistance. Among patients with frequent recurrences, continuous suppressive fluconazole therapy significantly reduces relapses and colonization, and it has tangential utility in preventing cryptococcosis and (in endemic areas) histoplasmosis, yet resistance remains a concern as it occurs with both intermittent and continuous therapy.⁶³

Final Key Questions

This evidence report addresses three issues -- risks to HIV-positive patients of invasive oral procedures, oral conditions as markers or indicators of change in HIV status (seroconversion and immunosuppression), and the efficacy or effectiveness of antifungal treatments (prophylactic and curative) for oral candidiasis in HIV-positive patients. As noted in Chapter 1, the second and third topics have been split to cover each of the aspects in parentheses; thus, we focus on five specific key questions.

Causal Pathways

Our causal pathways lay out the major elements of the key questions by specifying the study groups, variables, and outcomes of interest and the causal direction or links underlying them. Despite similarities between the two parts of Key Questions 2 and 3, each is represented in a separate causal pathway because of differences in the study groups.

Figure 1. Causal Pathway for Key Question 1: Are HIV/AIDS (more...)

   Figure 1. Causal Pathway for Key Question 1: Are HIV/AIDS Patients at Increased Risk of Complications from Intra-Oral Dental Procedures as Compared with Similar Patients without HIV/AIDS?

Figure 1

Figure 2. Causal Pathway for Key Question 2A: What (more...)

   Figure 2. Causal Pathway for Key Question 2A: What Are the Sensitivity, Specificity, and Positive and Negative Predictive Values of Oral Conditions as Markers of Recent HIV Seroconversion?

Figure 2

Figure 3. Causal Pathway for Key Question 2B: What (more...)

   Figure 3. Causal Pathway for Key Question 2B: What Are the Sensitivity, Specificity, and Positive and Negative Predictive Values of Oral Conditions as Indicators of Severe Immunosuppression?

Figure 3

Figure 4. Causal Pathway for Key Question 3A: How Efficacious (more...)

   Figure 4. Causal Pathway for Key Question 3A: How Efficacious Is Antifungal Drug Prophylaxis Against Oral Candidiasis Infection in Persons with HIV/AIDS?

Figure 4

Figure 4

Figure 5. Causal Pathway for Key Question 3B: How Effective (more...)

   Figure 5. Causal Pathway for Key Question 3B: How Effective Are Available Antifungal Drugs for the Treatment of Oral Candidiasis in Persons with HIV/AIDS?

Figure 5

Figure 4

Inclusion and Exclusion Criteria

The inclusion and exclusion criteria for the studies to be reviewed and abstracted evolved over time as we became more familiar with the relevant literature published on the five key questions. Fortunately, most of the literature was relatively recent (none published before the mid-1980s), so the article formats and materials presented were fairly standard, although far from completely. Reflecting the reality of the available articles, our original "ideal" standards on which studies to include were adjusted. This review represents the best of the research we identified to address these questions.

We tailored the criteria to the question as well as to the available literature. The basic screening criteria for inclusion and exclusion that applied to articles for all of the questions were as follows:

• Include: Only research manuscripts.

• Exclude: Literature reviews (except to check references for additional articles), letters, commentaries, editorials, clinical case reports, or practice or treatment guidelines.

• Include: Only articles published in English.

• Include: Only articles reporting on the human population.

For Key Question 1 on the differences in outcomes and complications of intra-oral procedures between HIV-positive and HIV-negative patients, the criteria used were as follows:

• Include: Studies reporting on complications of intra-oral surgical dental procedures, including orthognathic, periodontal, extractions, endodontics, prophylaxis, root planing and scaling, and implants.

• Exclude: Studies in which treatment is not rendered concurrently to HIV-positive and HIV-negative patients.

• Include: HIV status confirmed by test results.

• Exclude: Patient outcomes or complications not reported according to patient group and procedure.

• Include: Outcomes or complications reported are local infection, systemic infection, increased bleeding, dry socket (alveolitis), or delayed healing.

Because the nature of Key Questions 2A and 2B is so different from the nature of Key Question 1, which is more like a diagnostic question than an outcome difference question, the criteria for inclusion were different as well. For Key Question 2A on the use of oral lesions as markers for seroconversion, the criteria were as follows:

• Include: Reported presence of any of the following selected oral lesions within the 3-month primary HIV-infection window: hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, or parotid swelling.

• Include: Specificity, sensitivity, or positive or negative predictive values of the selected oral lesions as markers for seroconversion are reported, or information from which they can be calculated is presented.

• Exclude: No reasonable or firmly established time of exposure to HIV is reported.

The criteria for Key Question 2B addressing the use of selected oral lesions as indicators of change in immunosuppression were as follows:

• Include: Reported presence of any of the following oral lesions in HIV-positive patients: hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, parotid swelling, Kaposi's sarcoma, or linear gingival erythema.

• Include: HIV status of patients at time of oral lesion reported by CD4 cell count or plasma viral load.

• Include: Specificity, sensitivity, or positive or negative predictive values of the selected oral lesions as indicators of severe immunosuppression (CD4 <200 cells/mm³) are reported, or information from which they can be calculated is presented.

For Key Question 3A, addressing the use of available antifungal agents to prevent initial or recurrent oral candidiasis, the criteria were as follows:

• Include: Studies of the following available antifungal agents: nystatin formulations, clotrimazole, amphotericin B suspension, ketoconazole, fluconazole formulations, or itraconazole formulations.

• Include: Studies of the prevention of initial infection or recurrence of oropharyngeal candidiasis.

• Exclude: Studies without a concurrent control or comparison group.

• Include: Studies whose results are reported separately for initial infection or recurrence.

• Exclude: Studies on non-HIV/AIDS patients only or studies not reporting results separately for HIV/AIDS patients.

• Exclude: Studies without laboratory confirmation of oral candidiasis status at start of study period and after apparent infection.

• Exclude: Studies of mixed-site candidiasis where results are not reported separately for oropharyngeal candidiasis.

The criteria for Key Question 3B, on the effectiveness of available antifungal agents to treat oral candidiasis, were as follows:

• Include: Studies of treatment of oropharyngeal candidiasis.

• Include: Studies of available antifungal agents: nystatin formulations, clotrimazole, amphotericin B suspension, ketoconazole, fluconazole formulations, itraconazole formulations.

• Exclude: Studies comparing different formulations of the same drug.

• Exclude: Studies without a concurrent control or comparison group.

• Exclude: Studies on non-HIV/AIDS patients only or studies not reporting results separately for HIV/AIDS patients.

• Exclude: Studies without laboratory confirmation of oral candidiasis status at time of diagnosis and after treatment.

• Exclude: Studies of mixed-site candidiasis where results are not reported separately for oropharyngeal candidiasis.

Overview

Some experts have suggested that special provisions are necessary to avoid postoperative complications among HIV-infected patients receiving dental treatment.^{67 - 69} This suggestion is based on the rationale that because patients with HIV infection are immunologically compromised, they are at increased risk for complications associated with dental treatment. It follows that those with substantial immunosuppression (e.g., CD4 cell count <200 cells/mm³) or severe neutropenia (absolute neutrophil count <500 cells/mm³) are at greater risk for complications. However, the extent to which HIV-positive and HIV-negative individuals experience different rates of postoperative complications following routine dental procedures remains unclear.

Description of the Literature

Two of the seven dental procedures specifically mentioned to be of interest in Key Question 1 were the objects of study in the five articles. One article examined endodontic procedures (root canal treatment), and four studies examined tooth extraction. None reported on orthognathic surgery, periodontal surgery, prophylaxis, root planing and scaling, or implants.

The study of postoperative complications associated with endodontics employed a retrospective cohort design.⁷⁰ Patient records were used to determine the course of dental treatment, and related medical records confirmed HIV status. The sample included 48 adults, with a mean age of 34 years, who were predominantly male. The 32 HIV-positive patients were further categorized according to presence or absence of AIDS-related symptoms. Including both the HIV-positive and the HIV-negative groups, a total of 57 teeth underwent endodontic treatment. Nine of the HIV-positive patients and eight of the HIV-negative controls received antibiotics for relief of presenting symptoms before endodontic treatment. Complications were assessed at recall examinations for up to 3 months after treatment.

Four studies examined postoperative complications associated with tooth extraction. Two studies in this review employed retrospective cohort designs, and the other two used prospective cohort designs. The four studies followed a total of nearly 300 patients, the vast majority of whom were males. Across the four studies, the HIV-negative groups had low rates of prophylactic antibiotics (8 to 14 percent), and in the HIV-positive groups preoperative antibiotic coverage ranged from 24 to 63 percent.

Limitations in the Literature

One of our inclusion criteria required that studies used in this review have a concurrent HIV-negative control or comparison group receiving the same treatments in the same settings. Imposition of this criterion excluded studies¹⁸ that reported rather low rates of overall complications. Also excluded were studies that had a comparison group but that did not report complications separately for the two groups.⁷¹ The quality of the studies included in this review is variable. Only two studies employed a prospective study design. Two of the five studies looked at subgroups within the HIV-positive group. Only one study reported complication rates for subgroups under antibiotic therapy, and only one study of extraction complications included substantial numbers of female patients. Two of the five studies reported overall complication rates but no complication rates for individual types of complications. None controlled for severity of the presenting condition, and none adjusted for comorbid conditions, such as diabetes and periodontal disease. Finally, loss to follow-up was a concern in two studies.

Results for Endodontic Procedures

In the study of endodontics, the immediate (1- to 3-month follow-up) postoperative complication rate was exceedingly low in the HIV-positive group and nonexistent in the control group.⁷⁰ Only 1 of the 48 patients experienced any postoperative complications; an asymptomatic HIV-positive male was found to have pain and swelling following the initial root canal treatment. He received local debridement and antibiotics, and no further complications occurred. Otherwise, no complications associated with endodontic therapy were noted in any of the patients regardless of whether they did or did not receive prophylactic antibiotics. The authors did not detect a significant difference in complication rates between the two groups.

Results for Extractions

Four studies met the inclusion criteria to examine the issue of postoperative complications associated with tooth extraction. By way of background, previous studies that had reported postoperative complication rates with the removal of third molars in non-HIV-infected individuals demonstrated that complication rates could vary from 3 to 10 percent.⁷² Some evidence also suggests that the complication rates increase in patients with immunocompromising diseases.⁷³

Unadjusted complication rates overall ranged from 3 to 13 percent among HIV-negative subjects and 3 to 22 percent among HIV-positive patients. The study by Robinson and colleagues⁷⁴ suggested a trend toward lower incidence of dry socket (i.e., alveolitis) among those with antibiotic treatment, but the researchers found no substantial difference in complication rates between the HIV-positive and HIV-negative groups.

Porter and associates⁷⁵ reported no differences in delayed wound healing or postextraction dry socket. The only case of postextraction wound hemorrhage did occur in an HIV-infected patient but likely resulted from the patient's hemophilia. The authors concluded that postextraction complications are rare events in HIV-infected patients and do not recommend routine antibiotic prophylaxis before extractions.

In 1997, Dodson⁷⁶ reported finding no difference in complication rates between the HIV-positive and HIV-negative groups, with a relative risk (RR) of 1.68 and a 95 percent confidence interval (CI) of 0.82 to 3.62. In contrast, the 1994 Dodson et al. retrospective study⁷⁷ found that the groups differed significantly (RR = 7.04; 95 percent CI, 2.0 to 25.0; p = 0.001) and that the risk increased with increased levels of immunosuppression. When results were adjusted for age, preoperative antibiotics, and tobacco use, however, the difference was no longer statistically significant (RR = 4.3; 95 percent CI, 0.093 to 19.5; p = 0.06). Nevertheless, the authors suggested that biological and statistical evidence, while not overwhelming, does support the conclusion that HIV-positive patients are at increased risk of complications following extraction. They further asserted that because the complications are generally mild and managed conservatively, special measures such as determination of extent of immunosuppression or use of presurgical antibiotics are not recommended.

Three of the four studies in this review found no significant difference between complications in the HIV-positive and HIV-negative groups, although the HIV-positive groups tended to have more postoperative complications. The final study found that the HIV-positive groups had a statistically higher complication rate,⁷⁷ but with the risk factor adjustment it was no longer significant. Postextraction complications included persistent bleeding, persistent pain, localized alveolitis, local wound infection, and delayed wound healing. Nevertheless, across all studies, the postoperative complications were rather minor and were treated on an outpatient basis. Finally, based on their findings, none of the four studies called for the need to take special precautions in HIV-positive patients who do not have a coagulopathy (e.g., hemophilia, thrombocytopenia, or other known bleeding disorders) and are sufficiently healthy to be seen on an outpatient basis.

Overview

It has been observed that persons who have recently HIV seroconverted have a higher than expected prevalence of selected oral conditions, especially oral candidiasis and hairy leukoplakia.¹⁶ If the presence of these conditions among persons at high risk for becoming infected with HIV is an accurate marker for recent seroconversion, then much valuable time can be saved and effort can be directed at persons with those conditions to have them tested sooner, influence their behavior so as to prevent spreading the infection to others, and initiate therapy to stave off opportunistic infections and full-blown AIDS. It would be especially time- and resource-saving for clinical trials research and in the resource-poor nations of the world if such oral markers were so strong that they made further testing for HIV (such as using dual enzyme-linked immunosorbent assays [ELISAs] and a confirmatory Western blot) unnecessary. Of course, any oral conditions would have to be extremely highly associated with recent seroconversion to be considered a substitute for HIV testing because the result of being wrong could be devastating to an individual, risky for the public, and biasing for research.

The question has dictated that the medical "testing" dimensions of sensitivity (Sn), specificity (Sp), positive predictive value (PPV), and negative predictive value (NPV) of the oral condition as a marker of recent seroconversion be used as the way of evaluating its usefulness. As a marker, sensitivity in this context is an indicator of how accurately the presence of an oral condition in a group of persons predicts recent seroconversion. Specificity, on the other hand, reflects how accurately the absence of the oral condition predicts that recent seroconversion has not occurred in a group of persons. Persons who have the oral condition but are not HIV positive are false positives and diminish specificity, whereas those who are HIV positive but do not have the oral condition are false negatives and diminish sensitivity. Because the cost and consequences of false positives and false negatives can be substantial in both screening and research applications, an oral condition used in place of conventional HIV testing would need to have high sensitivity and moderate to high specificity.

In a clinical setting, however, where a healthcare provider is faced with a particular patient, the provider is in a different situation. The provider must decide, on the basis of the presence or absence of an oral condition, whether this patient is likely to have HIV seroconverted. Help answering this question is to be found in the predictive values for the oral condition. PPV is the probability that a patient with the oral condition has seroconverted, and NPV is the probability that a person without the oral condition has not seroconverted. However, positive and negative predictive values are affected by the prevalence in the study population of the disease being predicted - - in this case, recent seroconversion. Thus, positive and negative predicted values must be viewed in a more relative perspective than sensitivity and specificity (i.e., in the context of the prevalence of recent seroconversion in the population under study). If the prevalence of recent seroconversion is greatly different from population to population, positive and negative predictive values of the same magnitude may not have the same meaning to clinicians.

Description of the Literature

Results for Oral Candidiasis

The study by Colebunders and colleagues⁷⁸ used a prospective design to investigate the prevalence of a variety of signs and symptoms (including oral candidiasis) among two groups, each of 90 initially seronegative hospital patients in Zaire. All of the patients received blood transfusions for a variety of reasons between 1986 and 1987 before blood products were all routinely screened for HIV. The 90 "cases" consisted of all patients transfused with seropositive blood; the 90 "matched controls" received seronegative blood. More than 60 percent of the patients in the study were under 5 years of age. A large proportion of the patients in both groups were lost to follow-up and to death. No details were provided on the clinical determination of oral candidiasis. The controls were tested for HIV only at 12 months, and none of them had seroconverted, whereas 96 percent (all but two) of the available cases had seroconverted by the end of 3 months. While Sn, Sp, PPV, and NPV were not reported directly, data were reported that allowed us to calculate them.

The majority of patients with oral candidiasis had seroconverted by the end of 3 months (PPV = 82 percent), but only a small proportion of those who had seroconverted had oral candidiasis (Sn = 14 percent). Very few who did not seroconvert had oral candidiasis (Sp = 97 percent), and, as suggested above, most of those who did not have oral candidiasis did not seroconvert either (NPV = 57 percent).

Overview

Key Question 2B involves the presentation of evidence supporting the use of selected individual oral conditions as indicators of the progression of HIV infection to the stage of severe immunosuppression and a diagnosis of AIDS. As formulated, the question suggests the use of measures of the CD4 cell count as an indicator of immunosuppression. The question emerges from an interest in achieving earlier identification of immunosuppression in persons being treated for HIV for the purposes of making changes in their antiretroviral therapy. It also raises the possibility of oral conditions serving as an alternative end point in clinical trial research or as a substitute for the CD4 blood tests to identify immunosuppression in poor nations where medical and financial resources are scarce or nonexistent.¹⁶ As with the previous question on recent seroconversion, Sn, Sp, PPV, and NPV are suggested as the basis for assessing the usefulness of the oral conditions as indicators of immunosuppression.

The original question identified the following five oral conditions of interest: hairy leukoplakia, oral candidiasis, necrotizing ulcerative periodontitis, oral ulcers, and parotid swelling. Two additional relevant conditions of interest were added at the recommendation of the Technical Expert Advisory Group - - linear gingival erythema and Kaposi's sarcoma. Articles that evaluated the predictive capability of combinations of some or all of these conditions have not been included in this review because they were not consistent in the number, mixture, or specific types of conditions they included.^{79 - 81}

Description of the Literature

The 10 articles do not, however, represent 10 different study populations. There are two studies^84,85 from which much if not all of the data reported are the same as for two other studies.^51,52 Thus, the 10 articles represent only eight different study populations.

The study designs represented are a mixture of cross-sectional,^{18, 82} prospective cohort,^{47,50 - 52,83 - 85} and case-control.¹⁵ Only the data from the case portion of the case-control design is reported in the evidence table, and only data from the baselines of the prospective studies are included. This essentially reduces prospective designs to cross-sectional ones because the oral conditions and CD4 cell counts were assessed at the same or very similar points in time.

Nine out of 10 of the studies relied on clinical determination of candidiasis by means of an examination by dental or other health personnel (and sometimes by culture as well), and one of the articles was based solely on patient self-report of "thrush."⁸³ All but one of the nine clinical studies⁵⁰ mentioned training the examiners or reported having only one examiner.

Table 10. Articles in Evidence Table 3 Used to Report (more...)

Table 10. Articles in Evidence Table 3 Used to Report on Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value of Oral Conditions as Indicators of Severe Immunosuppression

Authors and Year	Conditions a
	OC	HL	NUP	OU	PS	LGE	KS
Feigal, Katz, Greenspan, et al., 199150	x	x		x			x
Barr, Lopez, Rua-Dobles, et al., 19925	x	x
Kolokotronis, Kioses, Antoniades, et al., 199447	x	x
Glick, Muzyka, Lurie, et al., 199418	x	x	x	x			x
Kirby, Munoz, Detels, et al., 199483	x
Begg, Panageas, Mitchell-Lewis, et al., 199652	x	x	x	x		x	x
Howell, Jandinski, Palumbo, et al., 199682	x
Phelan, Begg, Lamster, et al., 199785	x
Schuman, Sobel, Ohmit, et al., 199884	x
Schuman, Ohmit, Sobel, et al., 199851	x	x		x		x

a

OC = oral candidiasis, HL = hairy leukoplakia, NUP = necrotizing ulcerative periodontitis, OU = oral ulcers, PS = parotid swelling, LGE = linear gingival erythema, KS = Kaposi's sarcoma

The conditions on which the studies reported are quite varied. It should be noted that none of the articles reported on parotid swelling. However, all 10 articles reported on oral candidiasis, 6 on hairy leukoplakia, and 4 on oral ulcers. These were the most commonly studied conditions in the studies reviewed. Only two studies reported on linear gingival erythema, two on necrotizing ulcerative periodontitis, and three on Kaposi's sarcoma. One article alone⁵² reported on six of the oral conditions. There was one article that only reported data to compute the positive predicted value for oral candidiasis⁸⁵ and three other articles that reported on oral candidiasis alone.^{82 - 84}

A number of studies that are cited in the literature as providing support for a relationship between the presence of one of these oral conditions and CD4 level or viral load have not been included in the review. They have not been included because they do not report the sensitivity, specificity, positive predictive value, or negative predictive value (or the data necessary to calculate them), but rather report the mean or median CD4 or viral load levels^49,86,87 or a measure of association, such as an odds or likelihood ratio, without reporting the underlying distributions.^88,89

As noted above, all 10 articles reported one or more of the parameters (Sp, Sn, PPV, or NPV) or the data to calculate them for the relationship between an oral condition and CD4 cell count. The study populations are quite different, however. One is a sample composed mostly of congenitally infected young children.⁸² Another study group (two articles) is composed only of women,^51,84 three other studies are composed only of gay men,^15,50,83 and one other consists mostly of gay men.¹⁸ One of the studies (two articles) reports results separately for two different study groups, one composed of gay men (most of whom were white) and the other of male and female injected drug users (most of whom were black).^52,85 The studies also varied widely in the reported prevalence of oral candidiasis - - from 8 to 54 percent - - and in the prevalence of immunosuppression (CD4 <200 cells/mm³) - - from 13 to 53 percent.

The study patients were recruited from various settings. One study employed HIV-positive patients seeking dental care,¹⁸ others used patients in treatment for HIV,^47,82,83 and some recruited from community organizations¹⁵ or sampled communities expected to have a large number of HIV-infected persons.⁵⁰ The Schuman et al. study^51,84 reported that the sample was recruited partly from community and partly from medical care settings.

Limitations in the Literature

There are a number of obvious limitations associated with the evidence reviewed on oral lesions as indicators of immunosuppression in HIV-positive persons. First, the study populations in the studies reviewed may not be representative of the HIV-positive population generally. As discussed earlier, one study examined children only, two examined women only, and three examined only homosexual men. The prevalence of oral lesions and the prevalence of immunosuppression both varied widely among the studies. For clinical situations, the decision as to whether or not the presence of an oral lesion is a good indicator of a low CD4 cell count is based largely on the positive predictive value. However, the positive predictive value is a function of the prevalence of the disease in question, which in this case is immunosuppression as measured by CD4 cell count <200 cells/mm³. Because the prevalence of CD4 cell count may vary among different populations, results from these studies may not be generalizable to other populations. Second, the study by Glick et al.,¹⁸ a large study that reported on five of the oral conditions reviewed, examined HIV-positive patients referred for dental care. It is entirely possible that some of the patients went to the dentist because of the presence of oral lesions. If this occurred, it would likely increase the observed rate of true positives and false positives. Third, none of the studies reported on parotid swelling, a condition also suspected to be a consequence of immunosuppression, and none of the studies reported measured viral load. Fourth, the time interval between measurement of the CD4 cell count and the oral examination ranged from "within 24 hours" to "from 3 to 12 months" to not being reported. Long intervals may weaken or invalidate results. Because of new HIV highly active antiretroviral therapies (HAART) aimed at reducing viral load, viral load may be an even more important end point than CD4 cell count. Next, there was no agreement on the diagnostic criteria for the oral conditions to be included. In the case of oral candidiasis, for instance, they ranged from patient self-report to laboratory culture of Candida species to confirm a clinical diagnosis. Finally, although there were several studies investigating oral candidiasis and oral leukoplakia, there were far fewer investigating the other suspected indicators.

Results for Oral Candidiasis

Table 11. Summary of Sensitivity and Specificity in (more...)

Table 11. Summary of Sensitivity and Specificity in Articles Using Oral Candidiasis as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150					X													O
Barr, Lopez, Rua-Dobles, et al., 199215									X											O
Glick, Muzyka, Lurie, et al., 199418																X			O
Kirby, Munoz, Detels, et al., 199483							X													O
Kolokotronis, Kioses, Antoniades, et al., 199447														X						O
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a						X								O	X			O
Howell, Jandinski, Palumbo, et al., 199682											X						O
Phelan, Begg, Lamster, et al., 199785 MSM a IDU a	NR NR
Schuman, Sobel, Ohmit, et al., 199884									X								O
Schuman, Ohmit, Sobel, et al., 199851							X												O

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

The situation with respect to specificity is somewhat more consistent. The specificities range from 65 to 97 percent, a range of 32 percentage points, with a cluster of eight study groups in the upper end of the range -- from 79 to 97 percent. This suggests that only 3 to 35 percent of the persons with CD4 cell counts of 200 cells/mm³ or higher presented with oral candidiasis and were thus false positives for this oral condition as an indicator of immunosuppression.

Table 12. Summary of Positive Predictive Values and (more...)

Table 12. Summary of Positive Predictive Values and Negative Predictive Values in Articles Using Oral Candidiasis as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentagesb
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150									P									N
Barr, Lopez, Rua-Dobles, et al., 199215																PN
Glick, Muzyka, Lurie, et al., 199418															NP
Kirby, Munoz, Detels, et al., 199483									P										N
Kolokotronis, Kioses, Antoniades, et al., 199447																			PN
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a									P	P		N					N
Howell, Jandinski, Palumbo, et al., 199682													P			N
Phelan, Begg, Lamster, et al., 199785 MSM a IDU a								P	P
Schuman, Sobel, Ohmit, et al., 199884								P										N
Schuman, Ohmit, Sobel, et al., 199851								P										N

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

The range of NPVs is from 61 to 90 percent (a 29 percentage point spread) and considerably smaller than for PPVs. However, all but four of the NPVs are clustered between 84 and 90 percent. This suggests that from 10 to 39 percent of HIV-positive persons with oral candidiasis have CD4 cell counts of 200 cells/mm³ or higher and are not yet severely immunosuppressed.

Results for Hairy Leukoplakia

Six of the 10 studies reported on the association between hairy leukoplakia and CD4 cell count.^{15,18,47,50 - 52} One study⁵² reported sensitivity, specificity, and predictive values separately for homosexual men and intravenous drug users. The prevalence of hairy leukoplakia among the study populations ranged from 7 to 22 percent. The prevalence of CD4 cell counts <200 cells/mm³ ranged from 17 to 53 percent. Sensitivities, specificities, and predictive values were similar among the homosexual men and intravenous drug user groups reported in the Begg et al.⁵² study.

Table 13. Summary of Sensitivity and Specificity in (more...)

Table 13. Summary of Sensitivity and Specificity in Articles Using Hairy Leukoplakia as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150								X										O
Barr, Lopez, Rua-Dobles, et al., 199215						X												O
Glick, Muzyka, Lurie, et al., 199418						X													O
Kolokotronis, Kioses, Antoniades, et al., 199447					X													O
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSMa IDUa					X	X													O	O
Schuman, Ohmit, Sobel, et al., 199851				X																O

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Table 14. Summary of Positive and Negative Predictive (more...)

Table 14. Summary of Positive and Negative Predictive Values in Articles Using Hairy Leukoplakia as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150							P												N
Barr, Lopez, Rua-Dobles, et al., 199215									P						N
Glick, Muzyka, Lurie, et al., 199418												N			P
Kolokotronis, Kioses, Antoniades, et al., 199447							P									N
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a												P P	N			N
Schuman, Ohmit, Sobel, et al., 199851								P										N

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Results for Necrotizing Ulcerative Periodontitis

Two studies reported on necrotizing ulcerative periodontitis (NUP).^18,52 The study by Begg et al.⁵² reported results separately for homosexual men and intravenous drug users. In the study by Glick et al.,¹⁸ 67 percent of the subjects were male homosexuals/bisexuals. The study by Begg et al. reported only four cases of NUP among the 82 homosexual men and only four cases among the intravenous drug users. No cases of NUP were reported among the 87 intravenous drug users with a CD4 cell count <200 cells/mm³. In the study by Glick et al., 41 of the 454 patients had NUP. The prevalence of a CD4 cell count <200 cells/mm³ was 41 percent in the study by Begg et al. and 53 percent in the study by Glick et al.

Table 15. Summary of Sensitivity and Specificity in (more...)

Table 15. Summary of Sensitivity and Specificity in Articles Using Necrotizing Ulcerative Periodontitis as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Glick, Muzyka, Lurie, et al., 199418					X																O
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a	X			X																O	O

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Table 16. Summary of Positive and Negative Predictive (more...)

Table 16. Summary of Positive and Negative Predictive Values in Articles Using Necrotizing Ulcerative Periodontitis as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Glick, Muzyka, Lurie, et al., 199418												N								P
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a	P													N	N		P

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Results for Oral Ulcers

Four studies examined the association between oral ulcers and CD4 cell count.^{18,50 - 52} One study⁵² reported sensitivity, specificity, and predictive values separately for homosexual men and intravenous drug users. The prevalence of oral ulcers ranged from 2 to 18 percent in these studies. The prevalence of a CD4 cell count <200 cells/mm³ ranged from 17 to 53 percent.

Table 17. Summary of Sensitivity and Specificity in (more...)

Table 17. Summary of Sensitivity and Specificity in Articles Using Oral Ulcers as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150	X																				O
Glick, Muzyka, Lurie, et al., 199418			X																		O
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a			X		X														O	O
Schuman, Ohmit, Sobel, et al., 199851			X																		O

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Table 18. Summary of Positive and Negative Predictive (more...)

Table 18. Summary of Positive and Negative Predictive Values for Articles Using Oral Ulcers as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150	P																	N
Glick, Muzyka, Lurie, et al., 199418												N									P
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a								P				P		N		N
Schuman, Ohmit, Sobel, et al., 199851									P									N

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Results for Linear Gingival Erythema

Studies by Begg and colleagues⁵² and Schuman and colleagues⁵¹ examined the association between linear gingival erythema (LGE) and CD4 cell count. In the former study, results are reported separately for homosexual men and intravenous drug users; the prevalence of LGE was 13 percent among the homosexual men and 34 percent among the intravenous drug users. The prevalence was 14 percent among the patients (all women) in the study by Schuman et al.

Table 19. Summary of Sensitivity and Specificity in (more...)

Table 19. Summary of Sensitivity and Specificity in Articles Using Linear Gingival Erythema as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a					X			X						O					O
Schuman, Ohmit, Sobel, et al., 199851				X															O

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Table 20. Summary of Positive and Negative Predictive (more...)

Table 20. Summary of Positive and Negative Predictive Values in Articles Using Linear Gingival Erythema as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a					P						P	N		N
Schuman, Ohmit, Sobel, et al., 199851					P												N

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Results for Kaposi's Sarcoma

Three articles examined the relationship between Kaposi's sarcoma and CD4 cell count.^18,50,52 The study by Begg et al.⁵² reported results separately for homosexual men and intravenous drug users. Kaposi's sarcoma is a rare disease even among HIV-positive patients. The prevalence of Kaposi's sarcoma ranged from 0 to 7 percent. No cases were reported among the intravenous drug users in the study by Begg et al.

Table 21. Summary of Sensitivity and Specificity in (more...)

Table 21. Summary of Sensitivity and Specificity in Articles Using Kaposi's Sarcoma as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150			X																		O
Glick, Muzyka, Lurie, et al., 199418				X																	O
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a	n/a		X																		O n/a

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Table 22. Summary of Sensitivity and Specificity in (more...)

Table 22. Summary of Sensitivity and Specificity in Articles Using Kaposi's Sarcoma as an Indicator of CD4 Cell Count <200 cells/mm3

Authors and Year	Percentages b
	0	5	10	15	20	25	30	35	40	45	50	55	60	65	70	75	80	85	90	95	100
Feigal, Katz, Greenspan, et al., 199150																P		N
Glick, Muzyka, Lurie, et al., 199418																		N		P
Begg, Panageas, Mitchell-Lewis, et al., 199652 MSM a IDU a	n/a													N							P n/a

a

MSM = men who have sex with men; IDU = injected drug user

b

NR = not reported; X = sensitivity; O = specificity

Description of the Literature

Of the six clinical trials that evaluated the efficacy of antifungal medications used to prevent oral candidiasis in patients with HIV/AIDS, five were double-blind, placebo-controlled, randomized clinical trials,^{53,54,55,90,91} and one was an open-label, randomized clinical trial.⁹² The total sample size for each study ranged from a low of 14⁹¹ to a high of 323,⁵⁵ with participants drawn from both community clinic and hospital settings. Men made up the entire subject pool in four studies,^53,54,90,91 women were the entire subject population in one study,⁵⁵ and the remaining study did not report the subjects' gender.⁹² Level of immunosuppression and use of antiretroviral therapy were fairly similar between active and placebo groups, where reported.

Prophylactic drug treatment lasted from 12 weeks⁵⁴ to more than a year,⁵⁵ with follow-up evaluation intervals varying by study. End points for new oral candidiasis diagnosis typically required new signs and/or symptoms of oral candidiasis, confirmed by such microbiologic evidence as positive Candida culture, positive potassium hydroxide fungal smear (KOH prep), or additional colony-forming units in oral washings. One study merely reported prophylactic failures as "unsuccessful clinical response and infection recurred".⁹¹

One study⁵³ assessed the efficacy of two dose regimens of nystatin pastilles versus placebo, whereas varying doses of fluconazole were compared with placebo in the other five studies. Fluconazole dose regimens tested included 100 mg/week,⁵⁴ 150 mg/week,^90,91 200 mg/week,⁵⁵ 50 mg/day,⁹² and 100 mg/day.⁹² Four studies, in which all subjects had previously had oral candidiasis, evaluated prophylaxis against recurrence.^54,90-92 The remaining two studies evaluated antifungal efficacy for prevention of first infections in approximately half of the subjects and against recurrent oral candidiasis in the other half.^53,55

Limitations in the Literature

Several issues and limitations were identified in the literature to address the question of the efficacy of antifungal prophylaxis for oral candidiasis.

None of the studies reported any assessment of compliance with the prescribed study regimen in either the drug arm or the placebo arm. Compliance is an important issue because prophylaxis failure may be due to the patients' inability or unwillingness to adhere to the drug regimen rather than a true reflection of drug efficacy.

None of the trials evaluated the efficacy of antifungal prophylaxis among patients using the newest protease inhibitors, which contain the highly active antiretroviral therapy (HAART), because even the most recent study⁵⁵ was conducted during May 1992 and January 1994. During a 1-year follow-up in a case-control analysis of the effect of HAART on preventing oral candidiasis recurrences, oral candidiasis was diagnosed in 2 (7 percent) of 29 protease inhibitor-treated patients and 23 (36 percent) of 64 nonprotease inhibitor-treated patients (p < 0.001).⁹³ It has been suggested that in addition to their effect on improving immunocompetence, the protease inhibitors in these newer therapies may have in vitro and in vivo anticandidal activity.⁹⁴

Concern has arisen recently over the development of azole-resistant fungal strains as a consequence of the practice of chronic suppressive therapy. Few of the clinical trials involving fluconazole assessed the specific candidal species, such as proportion of C. albicans versus other non-C. albicans species, present in recurrent or new infections. Marriott and coworkers⁹⁰ reported no significant difference between non-C. albicans relapse in the fluconazole-treated patients (27 percent) compared with placebo-treated patients (13 percent). In the Just-Nubling and coworkers⁹² study, C. albicans grew from oral washings in all relapses, and C. tropicalis, C. glabrata, or C. krusei also occurred, yielding pathogen counts of >1,000 in five patients in the control group, three in the 50 mg/day fluconazole group, and four in the 100 mg/day fluconazole group.

No studies reported results of drug resistance testing for oropharyngeal candidal species in cases of prophylactic failure. Although Schuman and coworkers⁵⁵ provided vaginal speciation information and in vitro resistance data for vaginal species collected during a substudy examining in vitro resistance to fluconazole, neither species patterns nor their in vitro drug resistance characteristics were reported for oropharyngeal candidiasis recurrences.

No cost-effectiveness analyses for antifungal prophylaxis against oral candidiasis in HIV/AIDS patients were conducted. A recent analysis of the cost-effectiveness of preventing several AIDS-related opportunistic infections⁹⁵ suggested that 100 mg/day fluconazole (even in patients at higher risk, with CD4 cell counts of <50 cells/mm³) is generally not recommended for primary prophylaxis against more serious fungal infections, including esophageal and systemic candidiasis, cryptococcus, histoplasmosis, and coccidiomycosis. In this analysis, to achieve a cost-effectiveness threshold of $50,000 per quality-adjusted life-year saved, the cost of fluconazole ($2,372/year, according to the AIDS Cost and Services Utilization Survey) would have to be reduced more than 50 percent, to approximately $100 per month.

No studies were identified that met inclusion criteria that evaluated the efficacy of clotrimazole, amphotericin B suspension, ketoconazole, or itraconazole formulations as prophylactic measures for oral candidiasis in persons with HIV/AIDS.

Results for Antifungal Preventive Efficacy

MacPhail and coworkers⁵³ found that over a 20-week follow-up period, a multivariate proportional hazards model based on 121 subjects, 48 of whom developed oral candidiasis during follow-up, indicated that (after controlling for CD4 cell counts, Candida carriage, and history of clinical oral candidiasis), nystatin is effective in delaying time to oral candidiasis (p < 0.001). On average, one nystatin pastille a day reduced the hazard of oral candidiasis to 60 percent of the hazard of the placebo group. Two nystatin pastilles per day (400,000 U/day) reduced the hazard to 35 percent of the hazard of the placebo group.

Stevens and coworkers⁵⁴ assessed the effectiveness of 100 mg fluconazole per day compared with placebo in preventing recurrent oral candidiasis over 12 weeks among a small group of 25 men. They found that significantly fewer subjects on fluconazole (0 percent) than on placebo (61.5 percent) developed recurrences. Gastrointestinal and neurological side effects were slightly greater among fluconazole users than among the placebo group.

Marriott and coworkers⁹⁰ assessed the efficacy of 150 mg fluconazole per week over 6 months in the prevention of oropharyngeal candidiasis relapse after successful treatment of an acute episode among 84 men with moderate to severe HIV infection. Of 73 evaluable patients, the median time to relapse was >168 days for the fluconazole group and 37 days for the placebo group. For those completing 6 months of prophylactic treatment, significantly fewer in the fluconazole group (42 percent) relapsed, compared with those in the placebo group (96 percent). Adverse event rates were reported as being 0.3 adverse events per patient-month for fluconazole versus 0.2 adverse events per patient-month for placebo.

Leen and coworkers⁹¹ examined the effectiveness of 150 mg fluconazole per week over 24 weeks in the prevention of oropharyngeal candidiasis in a maintenance phase following a demonstrated clinical cure with 50 mg/day fluconazole over 14 to 28 days. Randomization to placebo or fluconazole maintenance occurred 7 days after treatment ended (with evidence of clinical cure but no report of mycological status) for the acute phase. Patients were requested to inform their clinician immediately if symptoms of fungal infection recurred and to return monthly for study visits. All patients on placebo had a recurrence, four within 3 weeks and one at 18 weeks. Of patients on fluconazole, two relapsed at 4 and 5 weeks, two withdrew at 16 weeks with no recurrence, and five were clinically successful, with recurrence prevented to the end of the study at 24 weeks. Fluconazole resulted in a significant reduction in relapse rate over placebo.

Schuman and coworkers⁵⁵ assessed the efficacy of 200 mg fluconazole weekly for a median of 17 months in the prevention of mucosal candidiasis in 323 women with HIV/AIDS. Although outcomes for vaginal and esophageal candidiasis were also included in this study, only the data for oropharyngeal candidiasis were reviewed. Weekly fluconazole was effective in preventing oropharyngeal candidiasis (relative risk [RR], 0.50; 95 percent confidence interval [CI], 0.33 to 0.74; p < 0.001). Confirmed oropharyngeal candidiasis recurrence occurred in 25.9 percent of the fluconazole group, compared with 42.2 percent of the placebo group. Infections that were considered confirmed or probably clinically resistant to fluconazole occurred in 3.7 percent of the fluconazole group and 4.3 percent of the placebo group. Of the 13 patients from both groups who had infections clinically resistant to fluconazole, 12 had oropharyngeal or esophageal candidiasis, with C. albicans isolated from 9 of 11 cultures that were available. Of the same 13 patients who developed resistant infections, 7 had CD4 cell counts <10 cells/mm³, 3 had CD4 cell counts between 10 and 50, and the remaining 3 had CD4 cell counts between 50 and 300. Although 25 percent of the patients receiving fluconazole and 14 percent of the patients on placebo had at least one adverse event, this was not significant. No specific toxicities were reported.

Just-Nubling and coworkers⁹² evaluated the efficacy in an open-label fashion of two different regimens of fluconazole (50 mg/day and 100 mg/day) in preventing oral candidiasis recurrences over a minimum of 6 months among 65 patients with advanced stages of HIV disease (CD4 <100 cells/mm³). Prophylaxis with fluconazole reduced the occurrence of oral candidiasis. In 20 (95.2 percent) of 21 placebo-treated patients, a total of 60 relapses occurred. In the 50 mg/day fluconazole group, 2 (11.1 percent) of 18 patients had four relapses. In the 100 mg/day fluconazole group, 4 (21.1 percent) of 19 had nine relapses. There were no significant differences in the relapse rate between the 50 and 100 mg/day dose regimens.

Description of the Literature

Of the 12 clinical trials that evaluated the effectiveness of antifungal medications used to treat oral candidiasis in patients with HIV/AIDS, all were drug comparison trials. Three were double-blind, randomized clinical trials;^96-98 5 were examiner-blinded;^{99 - 103} and 4 were open-label, randomized clinical trials, using no blinding.^{104 - 107} Total sample sizes ranged from a low of 37⁹⁷ to a high of 334,¹⁰¹ with participants drawn from a variety of settings, including private practice, community clinics, and hospitals. All studies but one,¹⁰⁷ which was conducted in a hospital in Zaire among AIDS patients, were undertaken in developed nations, including the United States, Canada, or Europe. Men and women were included in all studies where reported. Two studies^103,106 assessed drug efficacy in infants and children, whereas the remainder were conducted with adult populations. Levels of immunosuppression were fairly similar between groups, where reported.

Table 23. Randomized Clinical Trials of Antifungal (more...)

Table 23. Randomized Clinical Trials of Antifungal Medications Used to Treat Oropharyngeal Candidiasis in Patients with HIV/AIDS

Antifungal Drugs a
Authors and Year	Gentian Violet	Nystatin	Clotrimazole	Amphotericin B	Ketoconazole	Fluconazole	Itraconazole
Pons, Greenspan, Lozada-Nur, et al., 1997100		L				L
Flynn, Cunningham, Kerkering, et al., 1995103		L				L
Nyst, Perriens, Kimputu, et al., 1992107	L	L				X
Koletar, Russell, Fass, et al., 1990104			T			C
Pons, Greenspan, Debruin, 1993101			T				C
Murray, Koletar, Mallegol, et al., 1997102			T				L
De Wit, Weerts, Goossens, et al., 198997					C	C
Hernandez-Sampelayo 1994106					L	L
De Repentigny, Ratelle, Baril et al., 199698					C		C
Phillips, De Beule, Frechette, et al., 199896	C	L				C	L
Graybill, Vazquez, Darouiche, et al., 199899	C	L				C	L
De Wit, O'Doherty, De Vroey, et al., 1998108						X	X

a

C = capsule or table; L = liquid suspension or oral solution; T = troche (lozenge); X = unspecified oral dose.

Limitations in the Literature

Several issues and limitations were identified in the literature to address the question of the effectiveness of various antifungal drugs used in the treatment of oral candidiasis.

HIV disease status (clinical disease and level of immunosuppression) was not reported in approximately half of the studies.

Antiretroviral therapy use was reported in only two studies,^98,104 and in none involving the highly active antiretroviral therapy (HAART) protease inhibitors.

Assessment of compliance with the prescribed drug therapy was not universally undertaken and may have had an impact on effectiveness outcomes.

Several studies had low enrollment at randomization and subsequent attrition of subjects for clinical and mycological evaluation, creating underpowered studies and precluding statistical comparison of clinical and mycological cure rates between treatment arms. Relapse rate comparisons particularly may have been underpowered.

Speciation of candidal organisms in treatment failures was rare, and no drug resistance testing was undertaken. Flynn and coworkers¹⁰³ reported that fungal isolates were saved but no susceptibility testing was done. They also reported that the seven fluconazole recipients in their study involving infants and children with non-C. albicans isolates were less likely to have successful outcomes compared with those with only C. albicans isolates (p < 0.001). In this study, no C. krusei was isolated at baseline; however, three patients receiving fluconazole had this isolate in cultures after 7 days of treatment.

No cost-effectiveness analyses for antifungal treatment of oral candidiasis in HIV/AIDS patients were conducted, although Nyst and coworkers¹⁰⁷ report on costs associated with gentian violet, nystatin, and ketoconazole in Africa that could have a major impact on therapy selected.

No studies were identified that met inclusion criteria that evaluated the effectiveness of amphotericin B suspension or intravenous amphotericin B in persons with HIV/AIDS.

Nystatin Versus Fluconazole

Pons and coworkers¹⁰⁰ compared the efficacy of a 14-day regimen of either once daily 100 mg of fluconazole liquid suspension or four times daily 500,000 units of liquid nystatin in a randomized, examiner-blinded clinical trial involving 167 HIV-infected patients with oropharyngeal candidiasis. Fluconazole use resulted in significantly (p < 0.001) more clinical cures (87 percent vs. 5 percent), mycological eradication of Candida organisms (60 percent vs. 6 percent), and fewer relapses noted on day 28 (18 percent vs. 44 percent) than for nystatin users. Relapse rates were not significantly different by day 42 (fluconazole, 27 percent; nystatin, 11 percent). Subjects were matched for oropharyngeal candidiasis disease severity, but not HIV disease severity.

Flynn and coworkers¹⁰³ compared a single daily dose of fluconazole oral suspension (2 - 3 mg/kg per day after loading dose) with 400,000 units four times a day of nystatin over 14 days in the treatment of oropharyngeal candidiasis in 182 immunocompromised infants and children, ages 5 months to 14 years, in a 32-center, randomized, observer-blinded trial. Efficacy results for the 64 HIV-infected subjects were reported as a subgroup analysis. Among these patients, fluconazole was superior to nystatin in achieving clinical cure (80 percent vs. 29 percent, respectively; p < 0.001). Higher rates of mycological eradication were also found for fluconazole among the HIV-infected subgroup (83 percent vs. 5 percent for nystatin; p < 0.001).

Nystatin Versus Gentian Violet Versus Ketoconazole

Nyst and coworkers¹⁰⁷ conducted an open-label, randomized study among 141 AIDS inpatients in a hospital in Zaire, of whom 41.6 percent died of AIDS during the 14-day course of therapy for oropharyngeal or oropharyngeal and esophageal candidiasis, leaving 72 evaluable patients. After 14 days of therapy, oropharyngeal lesions had disappeared in similar portions of patients treated with 0.5 percent aqueous solution gentian violet mouth rinse twice daily (42 percent) and 200 mg/day oral ketoconazole (43 percent), and in a significantly (p < 0.05) lower portion of patients treated with 200,000 units of nystatin mouth rinse four times daily (9 percent). Microscopy was conducted with KOH prep at baseline and after 14 days of therapy, indicating significantly (p < 0.05) more microscopic Candida clearance among users of gentian violet (62 percent) and ketoconazole (57 percent) than nystatin (13 percent).

Clotrimazole Versus Fluconazole

Koletar and coworkers¹⁰⁴ reported on a small, randomized, open-label trial comparing the efficacy of 100 mg fluconazole daily and 10 mg clotrimazole five times daily for 14 days among 39 HIV-infected subjects for the treatment of oropharyngeal candidiasis. Clinical resolution at the end of treatment was significantly (p = 0.018) better among fluconazole users (100 percent) than clotrimazole troche users (65 percent). Additionally, fluconazole produced significantly (p = 0.004) greater mycological eradication rates (75 percent) than did clotrimazole (20 percent). Fluconazole-treated patients were more likely to remain disease-free during follow-up than those treated with clotrimazole (p = 0.014 at 28 days and p = 0.074 at 42 days); prolonged clinical responses were correlated with mycological eradication. This study was an earlier single-center report of outcomes from the multicenter study of the same design reported by Pons and coworkers¹⁰¹ described below.

Pons and coworkers¹⁰¹ reported on a prospective multicenter, randomized, examiner-blinded comparison of 14 days of fluconazole 100 mg capsule daily and clotrimazole 10 mg troche five times daily among 334 HIV-infected adults with oropharyngeal candidiasis. Among patients fairly well-matched for age, gender, race, baseline severity of oral discomfort, lesions and erythema due to oral candidiasis, and previous diagnosis of AIDS, both treatments were clinically effective at 14 days (91 percent clinical cure for fluconazole and 85 percent clinical cure for clotrimazole). Mycological outcome (eradication of Candida) was significantly better (p = 0.005) for fluconazole (65 percent) than for clotrimazole (48 percent). Among patients evaluable for relapse at day 28, significantly more (p < 0.001) patients on clotrimazole (50 percent) had relapsed compared with patients on fluconazole (18 percent). With fewer patients remaining to evaluate at day 42, 34 percent and 40 percent, respectively, of fluconazole and clotrimazole groups had relapsed. C. albicans was found in 98 percent of baseline cultures, compared with 93 percent of relapse cultures. Adverse event rates at 18 percent and 19 percent, respectively, for fluconazole and clotrimazole were comparable.

Clotrimazole Versus Itraconazole

Murray and coworkers¹⁰² reported on the results of a multicenter, examiner-blinded, randomized clinical trial of 14-day therapeutic courses of itraconazole oral solution 200 mg daily compared with clotrimazole troches five times daily for the treatment of oropharyngeal candidiasis in 162 immunocompromised patients, the majority of whom (123) were HIV-infected. Results are reported separately for the HIV-infected subjects. HIV/AIDS patients who received itraconazole oral solution had significantly greater rates of negative culture (64 percent vs. 29 percent, respectively) and clinical response plus negative culture (55 percent vs. 28 percent, respectively), compared with those who received clotrimazole (p < 0.01).

Ketoconazole Versus Fluconazole

De Wit and coworkers⁹⁷ conducted a double-blind, randomized comparison of the efficacy of 50 mg fluconazole daily versus 200 mg ketoconazole daily for 28 days to treat oropharyngeal candidiasis in a small population of 40 patients with AIDS-related complex (ARC) or an AIDS diagnosis in 1986 - 88. There was a significantly (p = 0.0445) better clinical response rate among patients on the fluconazole regimen (100 percent) than on the ketoconazole regimen (75 percent), although 46 percent of the fluconazole-treated patients relapsed within 30 days, compared with only 11 percent of ketoconazole-treated patients. Negative culture was obtained at the end of treatment in 87 percent of the evaluated patients in the fluconazole group and 69 percent of those evaluated in the ketoconazole group.

Hernandez-Sampelayo¹⁰⁶ evaluated the clinical and mycologic response of oral candidiasis to fluconazole oral suspension (3 mg/kg per day; mean, 14 days; range, 6 to 33 days) and ketoconazole oral suspension (7 mg/kg per day; mean, 16 days; range, 5 to 49 days) in 46 pediatric patients with either HIV or AIDS. This small, multicenter, open-label, randomized controlled trial suggested comparable efficacy and safety of the two drugs in children, although fluconazole users had higher clinical and mycological cure rates (88 percent and 71 percent, respectively), at the end of treatment over variable periods of 5 to 49 days, than those treated with ketoconazole (81 percent and 57 percent, respectively). By 4 weeks post-treatment, 50 percent of fluconazole users had relapsed, compared with 41 percent of ketoconazole users.

Ketoconazole Versus Itraconazole

De Repentigny and coworkers⁹⁸ compared the efficacy of a 200 mg capsule of itraconazole daily with a 100 mg tablet of ketoconazole daily for 14 days in the treatment of oropharyngeal and/or esophageal candidiasis in a randomized, double-blind, multicenter clinical trial in Montreal that enrolled HIV-infected adults from hospitals and private practices. Outcomes for oropharyngeal candidiasis were reported for 98 evaluable subjects. Oropharyngeal infection was clinically cleared in 21 days in marginally more (p = 0.061) subjects on itraconazole (71 percent) than on ketoconazole (60 percent). At 14 days, cultures were negative in 62 to 63 percent of both groups. Relapse rates were similar, and adverse events for both drugs were mild.

Itraconazole Versus Fluconazole

Phillips and coworkers⁹⁶ conducted a multicenter, double-blind, randomized clinical trial that enrolled 244 HIV-infected adults with oropharyngeal candidiasis to compare antifungal drug efficacy using three arms: 100 mg of itraconazole oral solution twice daily for 7 days, 100 mg of itraconazole oral solution once daily for 14 days, and 100 mg of fluconazole once daily for 14 days. Among evaluable patients, a complete response (clearance of all signs and symptoms) or marked improvement was noted in 90 percent of those on once-daily itraconazole and 90 percent of the fluconazole-treated individuals at the end of 14 days of treatment, compared with 82 percent of those on a 7-day regimen of twice daily itraconazole. This study demonstrated equivalence (p = 0.0024) in terms of clinical response between the 14-day courses of fluconazole and itraconazole. Drug tolerability and early (within 18 days) relapse (34 to 37 percent) were comparable between study groups.

Graybill and coworkers⁹⁹ reported on the outcomes of a multicenter, randomized, examiner-blinded equivalence trial among 179 HIV-infected patients with oropharyngeal candidiasis, treated in one of three arms: 200 mg of itraconazole oral solution daily for 7 days, 200 mg of itraconazole oral solution once daily for 14 days, and 100 mg of fluconazole once daily for 14 days. Both 14-day and 7-day regimens of itraconazole oral solution were equivalent to fluconazole for most efficacy parameters, with a clinical response rate (lesion eradication) of 97 percent for 14 days of itraconazole, 86 percent for 7 days of itraconazole, and 87 percent for 14 days of fluconazole. Approximately half of clinical responders in all groups had relapsed by 1 month after completion of treatment, with slight variations consistent with the drug regimen's ability to result in a culture negative status by the end of treatment.

De Wit and coworkers¹⁰⁸ compared the efficacy of a single dose of 150 mg of fluconazole with a 7-day regimen of 100 mg daily of itraconazole in the treatment of oropharyngeal candidiasis in an open-label, randomized trial of 40 HIV-infected adults. The fluconazole group produced significantly (p = 0.006) more subjects who were clinically cured (75 percent) or improved (15 percent) on day 8 than did the itraconazole group (24 percent cured, 12 percent improved). Mycological eradication was documented in only one subject from each treatment group at day 8. By the end of 30 days, 30 percent of the fluconazole group had relapsed, compared with 12 percent of the itraconazole group. The authors recognized that one of the major limitations of this study is that 50 percent of the patients on itraconazole were on concomitant medications that potentially affect this drug's bioavailability. Additional limitations include a small number of subjects, short therapeutic courses, the relatively low dose of itraconazole in capsule form, and assessment of clinical and mycological outcomes after a mere 8 days.

Good

The data are sufficient to assess the quality and strength of the findings. The outcome data are consistent and clearly support that the intervention, indicator, or drug in question does or does not demonstrate a significant or meaningful difference from the comparison group, gold standard, or alternative treatment/placebo.

Fair

The data are sufficient to assess the quality and strength of the findings. The outcome data are not consistent and do not uniformly/strongly support that the intervention, indicator, or drug in question does or does not demonstrate a significant or meaningful difference from the comparison group, gold standard, or alternative treatment/placebo, but the weight of the evidence is in one direction.

Poor

The data are not sufficient to assess the quality and strength of the findings. The outcome data are inconsistent, the study designs are not consistently strong, and there is no preponderance of evidence in one direction or the other to suggest whether or not the intervention, indicator, or drug is meaningfully different from the comparison group, gold standard, or alternative treatment/placebo.

Insufficient

The data are not sufficient to assess the quality and strength of any findings about the intervention, indicator, or drug in question because of a limited number of studies, small sample sizes, or poor study methodologies.

TEAG Members

• Dental Researcher

• Charles Barr, D.D.S., M.S.

• Division of Oral Medicine

• Department of Oral Diagnosis

• School of Dental Medicine

• University of Connecticut Health Center

• Clinical Experts

• John G. Bartlett, M.D.

• Chief, Division of Infectious Diseases

• School of Medicine

• Johns Hopkins University

• Michael Glick, D.M.D., B.MD.Sc.

• Programs for Medically Complex Patients

• University of Pennsylvania

• Lisa Kaplowitz, M.D.

• Director, HIV/AIDS Center

• Virginia Commonwealth University

• Consumer Organization

• Deborah Greenspan, B.D.S., L.D.S., R.C.S., D.Sc.

• Department of Stomatology

• School of Dentistry

• University of California - San Francisco

• Representing National Association of Persons with AIDS (NAPWA)

• User Group

• Sol Silverman, Jr., D.D.S., M.A.

• Department of Stomatology

• School of Dentistry

• University of California at San Francisco

• Representing American Dental Association (ADA)

Peer Reviewer List

• Clinical Experts

• Steven N. Abel, D.D.S.

• New York State Department of Health

• David Reznik, D.D.S.

• Oral Health Center, Infectious Diseases

• Grady Health System, Emory University

• Dental Professional Associations

• Miriam R. Robbins, D.D.S.

• Representative

• American Academy of Oral Medicine

• John A. Bartlett, M.D.

• Representative

• Infectious Disease Society of America

• Georgina P. Zabos, D.D.S., M.P.H.

• Representative

• American Public Health Association

• (APHA Oral Health Section)

• Patient, Consumer, and Public Health

• Charles Nelson*

• Representative

• National Association of People with AIDS

• Evidence Report Users

• Robert Maley

• Representative

• Health Partners of Minnesota

• Burt Spilker, M.D.*

• Representative

• Pharmaceutical Research and Manufacturers Association

*Reviewers who did not return comments.

Question 1: HIV Evidence Report Abstraction Form: Dental Treatment Complication Risk

Question 2A: HIV Evidence Report abstraction form: Sensitivity, Specificity, PPV, and NPV of Oral Lesions as Markers of seroconversion (Recent HIV infection 6-12 weeks, aka primary infection, acute HIV)

Question 2B: HIV Evidence Report Abstraction Form: Sensitivity, Specificity, PPV, and NPV of Oral Lesions for Immune Suppression and Viral Load

Question 3A: HIV Evidence Report Abstraction Form: Efficacy of Antifungal Prophylaxis For Oral Candidiasis in HIV Patients

Question 3B: HIV Evidence Report Abstraction Form: Effectiveness of Available Antifungal Agents for Treatment of Oral Candidiasis in HIV/AIDS patients

Research Design

Data Analysis

Measurement and the Validity of the Results

Quality Rating Scale for Studies of Oral Conditions as Markers of
Recent Seroconversion (Question 2A)
(score = sum of "yes" answer points multiplied by 5)

RESEARCH DESIGN

Measurement

Validity of Results

Quality Rating Scale for Studies of Oral Conditions as Indicators of
Severe Immune Suppression (Question 2B)
(score = sum of "yes" answer points multiplied by 5)

RESEARCH DESIGN

MEASUREMENT

VALIDITY OF RESULTS

Quality Rating Scale for Studies of Prophylactic Efficacy (Question 3A)
(score = sum of "yes" answer points multiplied by 4)

RESEARCH DESIGN

DATA ANALYSIS

MEASUREMENT AND THE VALIDITY OF THE RESULTS

Quality Rating Scale for Studies of Treatment Effectiveness
(Question 3B)
(score = sum of "yes" answer points multiplied by 4)

RESEARCH DESIGN